Research on dynamic torque control of hub motors based on model predictive control

As global energy transitions and carbon neutrality goals drive innovation, electric vehicles (EVs) are shifting from centralized to distributed drive architectures. In-wheel motors (IWMs), a key technology in distributed systems, are transforming vehicle design, energy management, and dynamic control through their integrated "motor-wheel"design. However, IWMs face challenges in complex road environments, including impact loads, temperature fluctuations, and electromagnetic interference. Traditional PID control struggles with parameter variations, load changes, and nonlinear friction, particularly in critical driving conditions like frequent starts/stops, regenerative braking, and torque vectoring. This study proposes a model predictive control (MPC) strategy to address these challenges. By integrating electromagnetic, mechanical, and control dynamics into a high-fidelity model, MPC predicts future motor states in real time and dynamically adjusts control inputs. Simulation results show that MPC significantly outperforms PI control in dynamic response, steady-state performance, and disturbance rejection. Specifically, MPC reduces torque overshoot from 180 Nm (PI control) to 3 Nm and minimizes steady-state fluctuations. These findings validate MPC's effectiveness in enhancing control precision, stability, and energy efficiency for IWMs, offering a strong foundation for advancing distributed drive architectures.